# Identifying and Combating Bias in Segmentation Networks by leveraging multiple resolutions

arXiv: https://arxiv.org/abs/2206.14919 · HTML: https://mlsanity.com/paper/2206.14919/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 23 references that are in the map.
- First posted on arXiv 29 Jun 2022.

## Builds on (most cited first, 11 of 11)

- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [Maxout Networks](https://mlsanity.com/paper/1302.4389/) (2013)
- [FastSurfer -- A fast and accurate deep learning based neuroimaging pipeline](https://mlsanity.com/paper/1910.03866/) (2019)
- [Fairness in Cardiac MR Image Analysis: An Investigation of Bias Due to Data Imbalance in Deep Learning Based Segmentation](https://mlsanity.com/paper/2106.12387/) (2021)
- [3D Whole Brain Segmentation using Spatially Localized Atlas Network Tiles](https://mlsanity.com/paper/1903.12152/) (2019)
- [Joint super-resolution and synthesis of 1 mm isotropic MP-RAGE volumes from clinical MRI exams with scans of different orientation, resolution and contrast](https://mlsanity.com/paper/2012.13340/) (2020)
- [Algorithm Fairness in AI for Medicine and Healthcare](https://mlsanity.com/paper/2110.00603/) (2021)
- [Detecting when pre-trained nnU-Net models fail silently for Covid-19 lung lesion segmentation](https://mlsanity.com/paper/2107.05975/) (2021)
- [FastSurferVINN: Building Resolution-Independence into Deep Learning Segmentation Methods -- A Solution for HighRes Brain MRI](https://mlsanity.com/paper/2112.09654/) (2021)
- [On the relationship between calibrated predictors and unbiased volume estimation](https://mlsanity.com/paper/2112.12560/) (2021)
- [Assessing Lesion Segmentation Bias of Neural Networks on Motion Corrupted Brain MRI](https://mlsanity.com/paper/2010.06027/) (2020)

## Abstract (arXiv)

Exploration of bias has significant impact on the transparency and applicability of deep learning pipelines in medical settings, yet is so far woefully understudied. In this paper, we consider two separate groups for which training data is only available at differing image resolutions. For group H, available images and labels are at the preferred high resolution while for group L only deprecated lower resolution data exist. We analyse how this resolution-bias in the data distribution propagates to systematically biased predictions for group L at higher resolutions. Our results demonstrate that single-resolution training settings result in significant loss of volumetric group differences that translate to erroneous segmentations as measured by DSC and subsequent classification failures on the low resolution group. We further explore how training data across resolutions can be used to combat this systematic bias. Specifically, we investigate the effect of image resampling, scale augmentation and resolution independence and demonstrate that biases can effectively be reduced with multi-resolution approaches.